Keeping Pace with Science The CyVerse Data Store in 2020 and the - - PowerPoint PPT Presentation

Keeping Pace with Science

The CyVerse Data Store in 2020 and the Future

Tony Edgin and Edwin Skidmore

iRODS UGM 2020

Data Store Statistics

★ 190 million data objects (9 PiB) ★ 80 million files (5 PiB) transferred in 2019 ★ 200 thousand files (14 TiB) transferred daily ★ 80 thousand users ★ 50 concurrent user connections on average

10 GiB File Transfer

Computation Platform Throughput (MiB/s) Texas Advanced Computing Center (TACC) 170 Jetstream 330 Amazon Web Services (AWS) 240 Google Cloud Platform (GCP) 260

File Transfer Performance Between CyVerse and Various Compute Platforms

What is the CyVerse Data Store?

• Offsite replication
• Optimization for accessing large sets of small files
• Event publishing
• Customer-driven extensions

○ Project-specific storage ○ Service integration (see Appendix) ○ Custom application integration (see Appendix)

Optimizing Access to Large Sets of Small Files

Use Case

Datasets for genome browser, e.g., JBrowse or UCSC Genome Browser

• thousands of kilobyte-sized files
• browsers are interactive,

○ loads files as needed ○ must be responsive, i.e., cannot take 20 seconds for each user request

CyVerse Solution

Set up a WebDAV server with a file cache

• apache web server with

○ davrods for iRODS access ○ modfilecache to cache files

• separate virtual hosts for anonymous and

authenticated access

• warm cache for byte-range access
• 100x faster than iget

Project-Specific Storage

Use Case

A project wants to store its data in the Data Store.

• 100 TB of data
• replicas stored locally at two institutions

CyVerse Solution

• project provides institutional storage servers
• CyVerse configures storage servers

○ catalog consumers hosting storage resources ○ project uses replication resource ○ policy to ensure data localities ○ separate iRODS service account

Steps toward utopia

• Increase interoperability
• Reduce accidental complexity

○ See Your app makes me fat

• Shorten scientific analysis feedback loop

Data Store of Tomorrow

Upcoming Features

• Thematic Real-time Environmental

Distributed Data Services (THREDDS) (see Appendix)

• Bring your own (BYO) infrastructure

○ BYO storage ○ BYO compute (later)

• Continuous analysis

User-Provided, S3-Compliant Storage

Use Case

User wants to analyze their cloud data using CyVerse cyberinfrastructure.

• data hosted in an S3-compliant storage

system, e.g. Google Cloud Storage

• moving them to Data Store is not feasible

CyVerse Solution

Use iRODS S3 Resource Plugin and Filesystem Scanner.

• cacheless, detached S3 resource for

scalability

• Filesystem Scanner registers data in place
• Filesystem Scanner runs on cloud platform to

avoid egress costs

• project owns cloud access credentials and

responsible for accrued costs More details in Appendix

CyVerse Continuous Analysis

Why “Continuous Analysis”?

• “Reproducibility of computational workflows is automated using continuous analysis”, CS

Greene et al, Nature Biotechnology, June 2016 (http://dx.doi.org/10.1038/nbt.3780) ○ Used github and drone to demonstrate “continuous analysis”, ci/cd for science ○ Code and data changes -> re-execute analysis and version everything ○ Authors admit limitations in dealing with data sets, though not impossible

• Scientists and researchers want event-driven analysis (data growth, sensors data, etc)
• Containers are becoming the de facto standard as units of reproducible compute
• Kubernetes is becoming the de facto standard for orchestrating containers
• Container orchestration and CI/CD technologies are difficult to use, esp for a scientist and

mortals who don’t know yaml (or json)

Why Continuous Analysis (cont.)

• Lessons learned

○ Jetstream/Atmosphere (multi-cloud, ad hoc interactive environments, allocations) ○ Containerized workflows ○ Data management

• Scientists need infrastructure to create, manage, and share in this emerging

Kubernetes-native analyses in a managed fashion

• Complements the CyVerse’s ecosystem, including Discovery Environment, Bisque, etc

Example User Stories

• I want my analyses to launch every time my workflow changes, my data changes, new

ML training data is available, or every hour

• I want my analyses to always be “available” and only be "charged" for the resources I

actually use

• I want to launch or transfer my analyses onto Jetstream/AWS/GCP/IoT/my own

project’s servers

• I want to use Argo, Airflow, Snakemake, or Makeflow workflows with Kubernetes and

scale as I define it

What is Continuous Analysis

Event-driven backend-as-a-service (BaaS) platform that will allow users to create, manage, deploy containerized analyses to any (kubernetes) cloud. High level Capabilities:

• Multi-cloud (and iRODS integrated)
• Auto-scaling and Scale to zero
• Event-driven aka Continuous analysis (CI/CD for science)

○ Data events, workflow events, periodic, external events

• Kubernetes/Cloud Native

○ Custom Resource Definition (CRD) ○ Supports k8s CRD workflows: standard k8s, Argo workflows

• Git for workflow persistence
• Support for federated identity (via keycloak)
• CyVerse-features: api, sharing/permissions, interop, etc

Current Status

• Currently, in development

○ REST API is the initial focus (not so easy) ○ Command line interface (somewhat easier) ○ Easy to use UI

• Limited release in Q4 2020

Questions?

Appendix

Service Integration

Use Case

A Powered by CyVerse service needs to access its users’ data.

• not controlled by CyVerse, no admin access

to data

• read-write access to its user data

CyVerse Solution

• service assigned rodsuser type iRODS

account

• user opts into service through User Portal
• shared collection for user and service

○

• wned by user in home collection

○ policy gives service write on contents ○ user has write permission, discourage delete, breaking service access

Example Application Integration

Use Case

Sparc’d is a desktop application supporting wildlife conservation created by Susan Malusa.

• manages sets of camera trap images

○ sizeable sets of small files ○ each set is tagged with metadata ○ supports sharing ○ images cannot be public, protect endangered species from poaching

• intended users are citizen scientists

○ volunteers, low frustration tolerance ○ require efficient uploads

CyVerse Solution

• project collection managed by Sparc’d creator

who gets own permission on contents, enforced by policy

• “tar pipe” style upload

○ Sparc’d packs images in one or more tar files ○ asynchronous rule unpacks, registers images

• metadata attached in bulk

○ uploaded as CSV in each tar file ○ applied by image registration rule

THREDDS Support for NetCDF Data Sets

Use Case

A project uses NetCDF files to store its public data sets.

• files are multi-gigabyte sized
• nly portions of some files needed at a time

CyVerse Solution

THREDDS Data Server (TDS) provides a collection

• f web services for accessing various types of

datasets including NetCDF.

• iRODS resource server and TDS share host
• TDS has direct, read-only access to iRODS

vault

• THREDDS data description files in vault
• project manages served data through iRODS
• analyst accesses data through TDS

THREDDS Integration Process

• 4. Analyst accesses

NetCDF data.

• 3. CyVerse sets up data residency

policy in iRODS and adds project to main TDS catalog. Project prepares data using iRODS client.

• 1. Project asks

for THREDDS integration.

TDS

iRODS

User-Specific S3 Resource Creation Process

User Data

core.re

User S3 Resource Discovery Environment Powered by CyVerse Services Continuous Analysis Platform

• 1. User gives

CyVerse the S3 connection information

S3

• 2. CyVerse

creates S3 resource and data residency policies.

• 3. User runs iRODS filesystem

scanner on cloud platform to register data.

• 4. User accesses data

from CyVerse platforms.

iRODS Filesystem Scanner

⟳